Fatty acid composition of Eucommia rubber particles, content and molecular weight distribution of Eu-rubber in Eucommia ulmoides leaves and samaras

: Abstact: [Backgroud] Eucommia ulmoides rubber particles (EuRPs) wrapped by a monolayer protein-lipid membrane are speculated to an important site for the biosynthesis of E. ulmoides rubber (Eu-rubber), and the types and contents of these protein and lipid varied in different E. ulmoides organs which might indicate the difference of Eu-rubber biosynthesis in EuRPs from different organs and tissues. In this paper, the morphological structure and fatty acid composition of EuRPs in leaves and samaras collected in October, 2019 were compared, and molecular weight (Mw) and molecular weight distribution (MWD) were also analyed by GPC system. [Results] The results showed that small rubber particles (SRPs, 1~4 μm) and medium rubber particles (MRPs, 4~7 μm) from leaves and samaras respectively accounted for 39%, 53% and 38%, 34%, while the amount of the large rubber particles (LRPs, 7~11 μm) were 23% and 13%, which indicated there was no significance in leaf and samara SRPs while the amount of MRPs and LRPs in leaf both were more than that of in samara. Synchronously, the determination results of fatty acids (FAs) compositon showed three of the 35 kinds FAs were uniquely detected in leaf rubber particles (LeRPs), which included eleven carbonic (C11:0), thirteen carbonic (C13:0) and arachidic acids (C20:0), while γ-linolenic acid (C18:3n6) was exclusively found in samara rubber particles (SaRPs) with a lower content. Additioanlly, five characteristic distribution peaks (CDPs) of Eu-rubber from leaves were detected while there were only 2-3 CDPs in samaras; Further, the MWD analysis suggested that the MWD of Eu-rubber in E. ulmoides samaras and leaves were 40.61 and 6.76 respectively, which indicated the more Eu-rubber molecules with various Mw existed in samara compared with those in leaf. [Conclusions] Together, we concluded that average size of LeRPs was bigger than SaRPs in October, and 3 kinds of saturated fatty acid (SAF) were unique to the leaves while only one of unsaturated fatty acid (UFA) was sole to SaRPs. Simultaneously, the Eu-rubber in samara in October possessed more rubber molecules with various molecular weights compared with the contemporaneous leaves.


Introduction
Eucommia ulmoides Oliver ('Duzhong' in Chinese; 'Tuchong' in Japanese), the single extant species of Eucommiaceae, is a dioecious, wind-pollinated tree diffusely distributed in the mixed mesophytic forest habitats of valleys, hills, and low mountains in central and eastern China (Cronquist 1981;Zhang et al. 2003). Duzhong is not only a valuably Chinese herbal medicines with the pharmacological functions to treat lumbago and hypertension, prevent bone senescence, and to enhance liver and kidney functions (Chinese Pharmacopoeia Committee 1990), but it has also been used as an important ru bber tree species (Bamba et al. 2001), especially as a natural substitute for -based synthetic rubber (PBSR). E. ulmoides rubber (EU-rubber), as a green renewable material, has been applied in the chemical industry medicine field as a raw material (Sun et al. 2013;Dong et al. 2011). EU-rubber is mainly composed of trans-1,4-polyisoprene (TPI), which is the isomer of natural rubber (NR), a cis-1,4-polyisoprene (CPI) (van Beilen and Poirier 2007;Zhang and Xue 2011). Previous studies reported rubber particles (RPs), including large rubber particles (LRPs) and small rubber particles (SRPs), were both the sites of NR biosynthesis (Nawamawat et al. 2011).
The results based on transmission electron microscopy (TEM) of tissue sections of Hevea brasiliensis shows that the particles consist of an apparently homogeneous core surrounded by a contiguous boundary, with the core appearing somewhat less electron-dense than the boundary, which implied the structure of the rubber (polyisoprene)-containing cores surrounded by biomembrane of non-rubber components (NRCs, e.g. proteins and phospholipids, etc) (de Fay et al. 1989;). Actually, the further studies proves that there are neutral lipids accounting for 2.4%, glycolipids and phospholipids for 1.0%, proteins for 2.2%, carbohydrates for 0.4%, ash for 0.2%, and other compounds for 0.1% in the commercial solid NR (Nishiyama et al. 1996). These NRCs are suspected of being responsible for both the superior properties and the inconsistent quality of NR (Chollakup et al. 2013). For instance, the whole lipid content have been reported to be inversely proportional to the tack properties of NR ) while unsaturated fatty acids (UFAs) were reported to act as a plasticizer for rubber by lowering the plasticity of NR ). On the contrary, some saturated fatty acids (SFAs) were found to affect NR crystallization by acting as a nucleating agent, which improved characteristic NR green strength in the unvulcanized state (Kawahara et al. 1999). As reported previously, the NR molecules chain consisting of ω-terminal and α-terminal, has been postulated to link with mono-phosphate and diphosphate groups associated with phospholipids by H-bonding at the α-terminal, and the ω-terminal is a dimethylallyl group linked to the protein by H-bonding . For the past few years, a series of evidence suggested that the ω-terminus bonds with a peptide or unidentified functional group, whereas the α-terminus consists of phosphoric ester and two long-chain fatty acid ester groups (Tarachiwin et al. 2005a, Tarachiwin et al. 2005b, Tarachiwin et al. 2005, which implys that the long chain fatty acid esters may be a constituent of the α-terminal group in NR chain that is expected to link with the rubber molecule as a component of a glyceride or phospholipid (Cornish 2001). Hence, the kind and content of fatty acids will be certainly to affect the properties of NR, e.g. molecular weight (Mw), molecular weight distribution (MWD), etc. Although, numerous studies around the biosynthesis of EU-rubber have been conducted, there are few reports on morphological structure and NRCs of E. ulmoides rubber paricles (EuRPs) because of the low extraction rate and complex steps of separation and purification of EuRPs. Therefore, it is of great significance to analyze NRCs (especially fatty acids, FAs) constitute of EuRPs and Mw of Eu-rubber, and it will provide a theoretical basis for further elucidating the morphological structure of EuRPs, proving the relationships of FAs content and kind with molecular characteristics of Eu-rubber.

Plant Materials and Reagents
Leaf and samara samples were collected randomly from 15 E. ulmoides clones (10 years old) with two or three plants per clone in a plantation of Guizhou University, and they were collected in October. EuRPs extracted from leaf or samara of E. ulmoides were stored in -80℃ for analysis of FAs and molecular weight of Eu-rubber. Trimethylolamine (Tris), phenyl-methylsulfonyl fluoride (PMSF), polyvinylpyrrolidone (PVP), β-mercaptoethanol (BME) and ascorbic acid (Vc) were purchased from Sigma-Aldrich LLC.; Potassium fluoride, magnesium sulfate, isopropanol and hydrochloric acid were purchased from Tianjin Kemiou Chemical Reagent Co., Ltd. Thirty-five standards of FAs were selected based on their different chain length, geometric structure (cis and trans) and double bond positions. The individual FAMEs were purchased from Fluka (>99%, Sigma-Aldrich). All solvents and reagents were of an analytical grade, specially for chromatography and purchased from Systerm (Systerm, Malaysia) except n-hexane, which was of a higher purity grade (Systerm, Malaysia, for GC, P99%). The esterifying agent TMS-DM (2 M) in n-hexane was purchased from Sigma (Sigma-Aldrich). Tetrahydrofuran (THF, chromatographically pure) and six mixed standards (PStQuick MP-M, Mw=706000, 96400, 5970, 474, 370 and 266 respectively) were purchased from TOSOH co. (Japan).

Extraction and purification of EuRPs
The EuRPs extraction were performed essentially as described by Yang et al. (2019) andZhao (1999) with some modifications. The fresh leaf or samara remaining in -80℃ ultra-low temperature freezer was rapidly grinded into powder and took 100 g of powder sample into a 1 L-flask containing 400 mL of the extract buffer [100 mmol·L -1 Tris-HCl, 50 mmol·L -1 KF, 2% (W/V) Vc, 5 mmol·L -1 MgSO4, pH=7.5). adding 2% (W/V) PVP before use, β-mercaptoethanol (final concentration is 5 mmol·L -1 ), 0.1 mmol·L -1 PMSF/ isopropanol solution] with rapid stirring extraction in the ice 1 min. After that, the filtrate from the sample homogenate through 200 mesh stainless steel screen under the condition of 4℃ was collected for later use. The filtrate was further divided into six centrifuge tubes (50 mL·tube -1 ) and the filtrate-containing tubes were centrifuged under the condition of 4℃ for 10 min at 5000 g with fixed-angle rotors (45 degrees) and the centrifuge tubes were gently took out. The liquid supernatant was orderly transferred to numbers of 10 mL centrifuge tubes by a 20 mL clean syringe for further centrifugation at 5000 g for 8 min below 4℃. The suspended solids on the top of liquid was then transferred to numbers of new 10 mL centrifuge tubes in which 8 mL of the rinsing solution was added in advance. The purification operation was performed by horizontal centrifugation of resuspension solution [100 mmol·L -1 Tris-HCl, 5 mmol·L -1 MgSO4, 10 mmol·L -1 β-mercaptoethanol] at 5000 g for 8 min under the condition of 4℃ and then remove rinsing solution, repeating the step of resuspendance and centrifugation 6-8 times. Purified EuRPs were stored in buffered 10% or 30% glycerol at -20℃ until prepared for microscopy.

Microscopic observation of EuRPs
Taking an appropriate amount (approx. 0.002 g) of EuRPs samples from E. ulmoides leaves and samaras in October on a clean glass slide, the samples were placed on a clean glass slide, and 1 drop of resuspension buffer was added to suspend the rubber particles. The cover glass was placed under Olympus BX43 optical microscope (Japan, 1000×) to observe the morphology of EuRPs. Meanwhile, the measuring system build-in OLYMPUS BX43 was used to measure the particle size of the EuRPs in the sample material collected from October, and the EuRPs observed were classified and counted according to the particle size. The shape and diameter of rubber particles were determined by scanning electron microscopy (SEM) method refered by Singh et al. (2003). A drop of a suspension of EuRPs in glycerol was fixed in 3% glutaraldehyde in 50 mM sodium cacodylate buffer (containing 1% tannic acid) for 1 h at room temperature. After washing in several changes of buffer, the samples were post-fixed in 1% aqueous osmium tetroxide also for 1 h at room temperature, and subsequently washed in several changes of distilled water. A small drop of the suspension (appropriately diluted) was air-dried on a piece of cover glass. The cover glass was then placed on a stub applied with a double-sided tape, sputter-coated with gold and examined with a Hitachi S530 SEM at 10 kV.

Determination of FAs in EuRPs
The extraction of FAs from EuRPs was conducted by referring to Kittigowittana et al. (2016) with some modifications. Take 80-100 mg ordinary sample of purified EuRPs into 15 mL centrifuge tube, and add 100 μL methyl nonadecanoate as internal standard. Continue to add 2 mL of 2% sodium hydroxide methanol solution for 30 min water bath at 85℃ in water bath kettle, add 3 mL 14% boron trifluoride (BF3)/methanol solution for treatment under the same condition. After the temperature was reduced to room temperature, extra 1 mL n-hexane was added into the centrifugal tube and the sample was extracted by shock for 2 min, then left for 1 h to wait for stratification. Take 100 μL of supernatant solution and allow to 1 mL with n-hexane. Use 0.45 μm filter membrane to pass the membrane and then test on the machine. In addition, take 30-50 mg of the grease sample and dissolve it in ethanol. And the processing steps are the same as above. After the extraction process, FAMEs were prepared before the GC analysis. One microlitre volume of each sample was injected into GC systems (Trace1310-ISQ, Thermo Fisher Scientific, USA) equipped with FID for separation and quantification of the FAMEs. The analysis was carried out using a TG-5MS fused silica capillary column (30 m, 0.25 mm i.d., 0.25 μm film thickness, Thermo Fisher Scientific, USA). The run was under an optimised temperature programme as follows: initial column temperature 80℃ for 1 min, programmed to increase at a rate of 10℃·min -1 up to 200℃ and then at 5℃·min -1 up to 250℃. Then at 2℃·min -1 up to final temperature of 270℃ and held for 3 min. Injector and detector temperatures were at 290℃. Helium as used as the carrier gas at a flow rate of 1.2 mL·min -1 with a split ratio of 30:1. The contents of fatty acids in the samples were calculated according to formula (1) :

Asi×mskli×Fj Astdi×m
Where Xi represents the amount of each fatty acid in the sample in milligrams per hundred grams (mg·Kg -1 ); Asi represents the peak area of fatty acids in the liquid was determined; mstdi represents the quality, in milligrams (mg), of the standard substance contained in the standard working solution of triglycerides of fatty acids extracted from the preparation of the standard determination solution; Fj represents conversion coefficients of triglycerides of fatty acids to fatty acids (see table 1); Astdi represents standard determination of the peak area of fatty acids in liquid; m represents the weight of the sample in grams (kg). The arithmetic mean of two independent measurements obtained under repeatable conditions was expressed, and three significant digits were reserved for recording.

Mw Determine of EU-rubber
Each EuRPs from leaves and samara were respectively dissolved in tetrahydrofuran (THF) to a concentration of 1.0 mg/mL, incubated at 50℃ overnight to allow for full dissolution and then filtered through a 0.22 μm polytetrafluoroethylene (PTFE) syringe filter. The molecular-weight and moleeuiar-weight distribution of EU-rubber were measured by gel permeation chromatography (EcoSEC system, TOSOH, Japan) equipped with an HLC-8320GPC controller, auto-injector (1 to 1, 500µL in 1µL increments), column oven (ambient plus 10°C to 60°C ), and an Bryce RI detector (tungsten light source, 1.00-1.80 RI range). The GPC was carried out using semi-micro SEC column (TSK-GEL SuperMultiporeHZ series) and PStQuick polystyrene standard kit was prepared as standard sample for minimum error. Sample volumes of 20 μL were passed through an TSK chromatographic column (4.6 IDmm × 150 mm, 3 μm particle size, no. 21488) with the column oven maintained at 40℃. The mobile phase was THF at a flow rate of 0.35 mL/min. The weight average molecular-weight and the molecular-weight distribution were obtained with EcoSEC-WS work station.

Statistical analysis
All results were obtained from three independent experiments and expressed as mean ± SD. Analyses of variance were perfonned by ANOVA procedures and significant differences (p < 0.05) were determined by T test with SPSS Statistics18.0.

Comparison of the morphological characteristics of EuRPs
The results of optical microscopy and scanning electron microscopy showed that the EuRPs were spherical and the shapes of the particles were different in different organs. In leaves of E. ulmoides, the content of small rubber particles (SRPs) (1-4 μm) and medium rubber particles (MRPs) (4-7 μm) were equal, accounting for 39% and 38% of the total EU-rubber particles respectively, and that of large rubber particles (LRPs) (7-11 μm) accounted for 23%. In the samara of E. ulmoides, the content of SRPs (1-4 μm) was the highest, accounting for 53%, the content of MRPs (4-7 μm) was 34%, and the content of LRPs was the lowest, accounting for 13% (Fig. 1). The above results indicated that the size of rubber particles in different organs and tissues of E. ulmoides showed high specificity, and the difference in particle size of this kind of rubber particles may be related to the synthesis of rubber molecules with different molecular weights. Under SEM condition, the morphology and size of leaf EuRPs were uniform and regular in October, while they were larger morphological difference in samara EuRPs (Fig. 2). Fig. 1 Optical microscope observation of EuRPs extracted from leaves (L) and samaras (S) L1, L2 or L3 S1 and S1, S2 or S3 represented for leaves rubber particles (LeRPs) and samara rubber particles (SaRPs) from 3 independent plants respectively

Analysis of Mw and MWD of Eu-rubber
The determination of molecular weight of Eu-rubber in E. ulmoides leaves and samara in October indicated that there were 5 peaks in the mass distribution of rubber molecules in Eucommia rubber particles in leaves, namely P1, P2, P3, P4 and P5, while the samara showed two to three peaks (P1, P2 or/and P3) (Fig. 3). Among them, the maximum molecular weight (Mw) of gutta-percha from Eucommia leaves rubber particles was 60.9 × 10 4 , followed by 1.61 × 10 4 , which were both significantly (p<0.05) or extremely significantly (p<0.01) higher than these of Eucommia samara (49.58×10 4 and 0.48×10 4 ). However, further determination of the average molecular weight among the peaks of Mw in gutta-percha respectively from E. ulmoides leaves and samara showed that the average molecular weight of gutta-percha in leaves was significantly lower (1.83×10 4 ) (P<0.05) than that in E. ulmoides leaves (4.15×10 4 ), which indicated that the latter contained the polymers with more high molecular weight than the former. Additionally, the analysis based on the molecular weight distribution (MWD, D = Mw/Mn) of E. ulmoides rubber found the D value of gutta-percha from E. ulmoides leaves and samaras were 6.76 and 40.61 respectively, which showed that rubber molecules with different molecular weight sizes in the leaves of E. ulmoides was more and concentrated distribution while rubber molecules with different molecular weight sizes in the samara of E. ulmoides was less and more dispersed.

Conclusions
E. ulmoides Oliver is a woody plant native to China which produces a hard, fibrous long-chain TPI (Nakazawa et al. 2009) and EuRPs (gutta-percha particles) are generally considered to be important sites for E. ulmoides rubber (Wuyun et al. 2018). In this study, we successfully separated EuRPs of E. ulmoides leaves and samaras in October and observed the morphology of gutta-percha particles in virtue of light microscope and electron microscope. The morphology observation of EuRPs extracted from Eucommia ulmoides samara and leaf found that there was little difference in EuRPs size of E. ulmoides leaves and the morphology was more uniform, among which 1-4 μm SRPs accounted for 39%, 4-7 μm MRPs accounted for 38%, and 7-11 μm LRPs accounted for 23%, while there was significant differences in EuRPs size of E. ulmoides samas, among which 1-4 μm SRPs accounted for 53%, 4-7 μm MRPs accounted for 34%, and 7-11 μm LRPs accounted for 13%. These results suggested that the difference in the distribution of different size particles in different organs and tissues may be related to the specific physiology and metabolism of these PRs in different tissue cells. Studies have shown that the RPs size in various organs of E. ulmoides was different in different periods and increased with the development of organs (Yang et al. 2019). On the one hand, the difference in EuRPs size might reflect differential accumulation of secondary metabolites (e.g. gutta-percha) in the these PRs during the growth and development of the plant organs contained gutta-percha; on the other hand, it also suggested that the difference in protein components (e.g., small rubber particle protein, SRPR) located in the surface of EuRPs which might be involved in the synthesis of gutta-percha with different molecular weight (Xiang et al. 2012;Jing, 2016).
Fatty acids analysis of leaf EuRPs collected in October by GC-MS method showed that a total of 30 fatty acids were detected in leaf EuRPs and SFAs and UFAs were accounted for 88.3% and 11.7% of the total fatty acids determined, respectively, and the results of ANOVA showed that the content difference between them reached a significant level (P<0.05), which indicated SFAs were the main FAs components in gutta-percha membrane of E. ulmoides leaf EuRPs collected in October. Further study found that the eleven carbonic acid (C11:0) and thirteen carbonic acid (C13:0) were the characteristic FAs in leaves EuRPs of E. ulmoides in October, while the γ-linolenic acid (C18:3n6) was a unique FA in samaras EuRPs of E. ulmoides. Moreover, the myristoleic acid (C14:1n5), elaidic acid (C18:1n9t) and arachidic acid (C20:0) of 35 fatty acids were detected neither in membrane of leaf EuRPs nor in membrane of samara EuRPs collected in October. Except for a few SFAs, e.g. twenty-one carbonic acid (C21:0), tricosanoic acid (C23:0) and twenty-four carbonic acid (C24:0), the contents of many UFAs including palmitoleic acid (C16:1n7), heptadecanoic acid (C17:1n7), oleic acid (C18:1n9c), α-linolenic acid (C18:3n3), eicosadienoic acid (C20:2), eicotrienoic acid (C20:3n6), docosadienoic acid (C22:2n6) and twenty-two carbohexanthin acid (C22:6n3) in plasmalemma of EuRPs isolated from samara were very significant (p<0.01) higher than these in leaf EuRPs, and the composition of these significantly different fatty acids might indicate the specific function of samara EuRPs in the synthesis of gutta-percha molecules. It is generally believed that UFAs can be inserted into the lipid layer of the plasmalemma, participate in its composition, affect the stability and fluidity of the membrane structure, and are important factors determining the function of the biological membrane (Li et al. 2009). The study found that the viscosity of NR was inversely proportional to its lipid content .  reported the UFAs could reduce the plasticity of NR by being used as plasticizers, on the contrary, some SFAs could influence the crystallization of NR as nucleating agents, and nucleating agents could improve the strength of NR in the unvulcanized state (Kawahara et al. 2011).
The Mw and MWD of Eu-rubber are closely related to the physical properties of gutta-percha, which is an important basis for determining the application range of Eu-rubber (Yang et al. 2019). The use of Eu-rubber with high Mw is equivalent to that of NR and can be used as a "green tire" and the Eu-rubber with low Mw is characterized by low melting point and strong hardness, which can be used as medically functional materials, shape memory materials and additives (De Barros et al. 1992;Kovuttikulrangsie and Sakdapipanich, 2005). Additionally, Those Eu-rubber with the same Mw and a narrower MWD have greater strength and plasticity, while other gutta-percha with wider MWD is better elastic, more wear-resistant and heat-resistant (Westall 1968;Guo et al. 2015). GPC analysis of MWD of Eu-rubber molecules in E. ulmoides leaf collected in October showed that there were five characteristic peaks, including high molecular weight (HMw), medium molecular weight (MMw) and low molecular weight (LMw), while MWD of Eu-rubber molecules in E. ulmoides samara collected at the same month contained two or three characteristic peaks. The average HMw of Eu-rubber molecules from leaf EuRPs and samara EuRPs were 60.9×10 4 and 49.58×10 4 , and the HMw in the former was significantly (p<0.05) higher than that in the latter. Moreover, the secondary HMw of Eu-rubber in leaf EuPRs was 1.61×10 4 , which it was extremely significant (p<0.01) higher than that of samara EuPRs (0.48×10 4 ). Among the 5 peaks of MWD of Eu-rubber from leaf EuPRs, the widest peak area was observed in secondary HMw of gutta-percha when it appeared in secondary HMw or third HMw of gutta-percha from samara EuPRs. Compared the secondary HMw or third HMw of MWD of Eu-rubber molecules from samara, the peak area of secondary HMw of Eu-rubber from leaf was much larger than the former, which indicated the latter had more gutta-percha molecules with different molecular weights. In leaf EuRPs, the average weight-average molecular weight (Mw ̅̅̅̅̅ ) of all 5 characteristic peaks of Eu-rubber molecules was 1.83×10 4 while the average number-average molecular weight (Mn ̅̅̅̅ ) was 0.30×10 4 , and the value of MWD was the ratio of Mw ̅̅̅̅̅ to Mn ̅̅̅̅ (6.1). Synchronously, in samara EuRPs, there were two kinds of average weight-average molecular weight (Mw ̅̅̅̅̅ ) of all 2-3 characteristic peaks of Eu-rubber molecules, namely Mw1 ̅̅̅̅̅̅̅ and Mw2 ̅̅̅̅̅̅̅ , and there accordingly were two kinds of the average numberaverage molecular weight (Mn1) ̅̅̅̅̅̅̅ and (Mn2 ̅̅̅̅̅̅ ). Therefore, the MWD1 and MWD2 were 27.3 and 54.3, respectively. Whatever MWD1 or MWD2 of Eu-rubber in samara EuRPs, both were significant larger than MWD of Eu-rubber in leaf EuRPs (6.1), and the results suggested that there was a larger molecular weight difference of gutta-percha molecules from samara EuRPs while there was a smaller one in gutta-percha molecules from leaf EuRPs. In brief, the Eu-rubber extracted from E. ulmoides leaves in October with characteristics of uniform molecular weight distribution was facilitated to control quality of Eu-rubber and the counterpart from E. ulmoides samaras had excellent processing property but poor uniformity.
Ethics approval and consent to participate I certify that the substance of this thesis has not been previously submitted for any other journals. Also, I certify that to the best of my knowledge any help received in preparing this paper and all sources used, have been ackonwledged in this paper.